Safety in bad weather

The functionality of future driver assistance, automation and safety functions must be permanently in place to ensure support for the driver and protection for passengers and other road users. The predictive vehicle sensor technology must therefore reliably and robustly detect its environment under all environmental conditions.

In C-ISAFE, environmental influences such as rain, fog and sun levels are simulated in the indoor test facility with specially developed test systems for weather as well as light. This allows weather effects and influences on the sensors to be researched and tested.

Influence of weather

To increase safety and comfort through driver assistance systems up to automated driving, the reliable use of environmental sensor technology is mandatory under all environmental conditions. In C-ISAFE, the performance of sensor systems is tested under reproducible boundary conditions in a defined test environment by using weather systems. For the different sensor types camera, radar and LiDAR, different kinds of disturbance effects caused by rain and fog are determined and characterized. Methods for the reduction of disturbance effects on sensor systems as well as models for (sensor) disturbance variables are developed in order to enable an optimal environment perception.

In addition to investigating and optimizing the performance of individual sensor types, C-ISAFE is particularly investigating the fusion of radar, camera, and LiDAR data to achieve the robust deployment of predictive sensor technology in weather conditions. The different types of sensor technologies have different strengths and weaknesses, and through intelligent combinatorial use, there is the potential to realize trustworthy object detection even in difficult conditions. For this purpose, new approaches combine the different sensor technologies using multimodal AI fusion algorithms.

Influence of light

A particular problem in the use of cameras is the contrast behavior in different light intensities. Challenging exposure situations are, for example, sunrise or headlights in oncoming traffic, since overexposed and underexposed areas lead to disturbances in the perception of the surroundings by driver assistance and highly automated systems. For example, backgrounds overexposed by the sun lead to low contrasts between objects within the image, or reflective surfaces blind the optical sensors of the systems. As a result, objects are incorrectly detected or not detected, resulting in system failures or faulty system activities. In order to investigate such effects with the optical sensors camera and LiDAR in a repeatable and targeted manner, C-ISAFE is developing a flexible light simulator. This is freely adjustable in height and angle of incidence. Together with the different lighting types, especially red light for low sun and white light for LED headlights, a variety of lighting scenarios can be simulated. In this way, reproducible boundary conditions are created in a defined test environment for the systematic investigation of optical sensor systems.

Ongoing projects

MIAMy

Accelerate Market Introduction of Autonomous Mobility

KICSAFe

KI-basierte Crasherkennung für das Sichere Automatisierte Fahren

Finished projects

SAFIR

Safety for all – Innovative Research Partnership on Global Vehicle and Road Safety Systems

SAVE-ROAD

Safe vision-based estimation of crash severity and reference system for autonomous driving

SimuSens

Entwicklung eines mobilen Prüfstands für optische Sensoren von Fahrerassistenz- und Sicherheitssystemen

TEPS

Test und Entwicklung einer passiven Sicherheitsfunktion

Publications

Since 2020
  • R. Huber, K. Schneider, A. Wetzel, E. Neitzel, and T. Brandmeier, "Light Analysis for Optimized Object Detection with Cameras for Integrated Safety Systems," in 2022 International Conference on Electrical, Computer and Energy Technologies (ICECET), Prague, Czech Republic,2022, pp. 1-6, doi: 10.1109/ICECET55527.2022.9872761.
  • D. Steinhauser, P. Held, B. Thöresz, and T. Brandmeier, "Towards Safe Autonomous Driving: Challenges of Pedestrian Detection in Rain with Automotive Radar," in 2020 17th European Radar Conference (EuRAD), Utrecht, Netherlands, 2021, pp. 409-412, doi: 10.1109/EuRAD48048.2021.00110.
  • D. Vriesman, B. Thoresz, D. Steinhauser, A. Zimmer, A. Britto, and T. Brandmeier, "An Experimental Analysis of Rain Interference on Detection and Ranging Sensors," in 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Rhodes, Greece, 2020, pp. 1-5, doi: 10.1109/ITSC45102.2020.9294505.
Before 2020
  • D. Schöppe, D. Steinhauser, B. Thöresz, S. Hasirlioglu, and T. Brandmeier, "Behavior of sensor systems for safety in automated driving with different weather conditions under reproducible conditions,” in 12. VDI-Tagung Fahrzeugsicherheit im Umfeld von neuen Rating- und Gesetzesanforderungen, 2019, pp. 205 – 218, doi: 10.51202/9783181023648-205.

Contact

Scientific Director CARISSMA – ISAFE, Research Professor for Vehicle Safety and Vehicle Mechatronics
Prof. Dr.-Ing. Thomas Brandmeier
Phone: +49 841 9348-3840
Room: H023
E-Mail:
Research Assistant C-ISAFE
Dr. rer. nat. Dagmar Steinhauser
Phone: +49 841 9348-3375
Room: H120
Fax: +49 841 9348-993375
E-Mail: